llvm-project/llvm/lib/Target/PowerPC/PPCInstrInfo.h

283 lines
12 KiB
C
Raw Normal View History

//===-- PPCInstrInfo.h - PowerPC Instruction Information --------*- C++ -*-===//
//
// The LLVM Compiler Infrastructure
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
//
// This file contains the PowerPC implementation of the TargetInstrInfo class.
//
//===----------------------------------------------------------------------===//
#ifndef LLVM_LIB_TARGET_POWERPC_PPCINSTRINFO_H
#define LLVM_LIB_TARGET_POWERPC_PPCINSTRINFO_H
#include "PPC.h"
#include "PPCRegisterInfo.h"
#include "llvm/Target/TargetInstrInfo.h"
#define GET_INSTRINFO_HEADER
#include "PPCGenInstrInfo.inc"
namespace llvm {
/// PPCII - This namespace holds all of the PowerPC target-specific
/// per-instruction flags. These must match the corresponding definitions in
/// PPC.td and PPCInstrFormats.td.
namespace PPCII {
enum {
// PPC970 Instruction Flags. These flags describe the characteristics of the
// PowerPC 970 (aka G5) dispatch groups and how they are formed out of
// raw machine instructions.
/// PPC970_First - This instruction starts a new dispatch group, so it will
/// always be the first one in the group.
PPC970_First = 0x1,
2010-12-24 12:28:06 +08:00
/// PPC970_Single - This instruction starts a new dispatch group and
/// terminates it, so it will be the sole instruction in the group.
PPC970_Single = 0x2,
/// PPC970_Cracked - This instruction is cracked into two pieces, requiring
/// two dispatch pipes to be available to issue.
PPC970_Cracked = 0x4,
2010-12-24 12:28:06 +08:00
/// PPC970_Mask/Shift - This is a bitmask that selects the pipeline type that
/// an instruction is issued to.
PPC970_Shift = 3,
PPC970_Mask = 0x07 << PPC970_Shift
};
enum PPC970_Unit {
/// These are the various PPC970 execution unit pipelines. Each instruction
/// is one of these.
PPC970_Pseudo = 0 << PPC970_Shift, // Pseudo instruction
PPC970_FXU = 1 << PPC970_Shift, // Fixed Point (aka Integer/ALU) Unit
PPC970_LSU = 2 << PPC970_Shift, // Load Store Unit
PPC970_FPU = 3 << PPC970_Shift, // Floating Point Unit
PPC970_CRU = 4 << PPC970_Shift, // Control Register Unit
PPC970_VALU = 5 << PPC970_Shift, // Vector ALU
PPC970_VPERM = 6 << PPC970_Shift, // Vector Permute Unit
PPC970_BRU = 7 << PPC970_Shift // Branch Unit
};
} // end namespace PPCII
2010-12-24 12:28:06 +08:00
class PPCSubtarget;
class PPCInstrInfo : public PPCGenInstrInfo {
PPCSubtarget &Subtarget;
const PPCRegisterInfo RI;
bool StoreRegToStackSlot(MachineFunction &MF,
unsigned SrcReg, bool isKill, int FrameIdx,
const TargetRegisterClass *RC,
SmallVectorImpl<MachineInstr*> &NewMIs,
bool &NonRI, bool &SpillsVRS) const;
bool LoadRegFromStackSlot(MachineFunction &MF, DebugLoc DL,
unsigned DestReg, int FrameIdx,
const TargetRegisterClass *RC,
SmallVectorImpl<MachineInstr*> &NewMIs,
bool &NonRI, bool &SpillsVRS) const;
virtual void anchor();
protected:
/// Commutes the operands in the given instruction.
/// The commutable operands are specified by their indices OpIdx1 and OpIdx2.
///
/// Do not call this method for a non-commutable instruction or for
/// non-commutable pair of operand indices OpIdx1 and OpIdx2.
/// Even though the instruction is commutable, the method may still
/// fail to commute the operands, null pointer is returned in such cases.
///
/// For example, we can commute rlwimi instructions, but only if the
/// rotate amt is zero. We also have to munge the immediates a bit.
MachineInstr *commuteInstructionImpl(MachineInstr *MI,
bool NewMI,
unsigned OpIdx1,
unsigned OpIdx2) const override;
public:
explicit PPCInstrInfo(PPCSubtarget &STI);
/// getRegisterInfo - TargetInstrInfo is a superset of MRegister info. As
/// such, whenever a client has an instance of instruction info, it should
/// always be able to get register info as well (through this method).
///
const PPCRegisterInfo &getRegisterInfo() const { return RI; }
ScheduleHazardRecognizer *
CreateTargetHazardRecognizer(const TargetSubtargetInfo *STI,
const ScheduleDAG *DAG) const override;
ScheduleHazardRecognizer *
CreateTargetPostRAHazardRecognizer(const InstrItineraryData *II,
const ScheduleDAG *DAG) const override;
[PowerPC] Fix the PPCInstrInfo::getInstrLatency implementation PowerPC uses itineraries to describe processor pipelines (and dispatch-group restrictions for P7/P8 cores). Unfortunately, the target-independent implementation of TII.getInstrLatency calls ItinData->getStageLatency, and that looks for the largest cycle count in the pipeline for any given instruction. This, however, yields the wrong answer for the PPC itineraries, because we don't encode the full pipeline. Because the functional units are fully pipelined, we only model the initial stages (there are no relevant hazards in the later stages to model), and so the technique employed by getStageLatency does not really work. Instead, we should take the maximum output operand latency, and that's what PPCInstrInfo::getInstrLatency now does. This caused some test-case churn, including two unfortunate side effects. First, the new arrangement of copies we get from function parameters now sometimes blocks VSX FMA mutation (a FIXME has been added to the code and the test cases), and we have one significant test-suite regression: SingleSource/Benchmarks/BenchmarkGame/spectral-norm 56.4185% +/- 18.9398% In this benchmark we have a loop with a vectorized FP divide, and it with the new scheduling both divides end up in the same dispatch group (which in this case seems to cause a problem, although why is not exactly clear). The grouping structure is hard to predict from the bottom of the loop, and there may not be much we can do to fix this. Very few other test-suite performance effects were really significant, but almost all weakly favor this change. However, in light of the issues highlighted above, I've left the old behavior available via a command-line flag. llvm-svn: 242188
2015-07-15 04:02:02 +08:00
unsigned getInstrLatency(const InstrItineraryData *ItinData,
const MachineInstr *MI,
unsigned *PredCost = nullptr) const override;
Improve instruction scheduling for the PPC POWER7 Aside from a few minor latency corrections, the major change here is a new hazard recognizer which focuses on better dispatch-group formation on the POWER7. As with the PPC970's hazard recognizer, the most important thing it does is avoid load-after-store hazards within the same dispatch group. It uses the POWER7's special dispatch-group-terminating nop instruction (instead of inserting multiple regular nop instructions). This new hazard recognizer makes use of the scheduling dependency graph itself, built using AA information, to robustly detect the possibility of load-after-store hazards. significant test-suite performance changes (the error bars are 99.5% confidence intervals based on 5 test-suite runs both with and without the change -- speedups are negative): speedups: MultiSource/Benchmarks/FreeBench/pcompress2/pcompress2 -0.55171% +/- 0.333168% MultiSource/Benchmarks/TSVC/CrossingThresholds-dbl/CrossingThresholds-dbl -17.5576% +/- 14.598% MultiSource/Benchmarks/TSVC/Reductions-dbl/Reductions-dbl -29.5708% +/- 7.09058% MultiSource/Benchmarks/TSVC/Reductions-flt/Reductions-flt -34.9471% +/- 11.4391% SingleSource/Benchmarks/BenchmarkGame/puzzle -25.1347% +/- 11.0104% SingleSource/Benchmarks/Misc/flops-8 -17.7297% +/- 9.79061% SingleSource/Benchmarks/Shootout-C++/ary3 -35.5018% +/- 23.9458% SingleSource/Regression/C/uint64_to_float -56.3165% +/- 25.4234% SingleSource/UnitTests/Vectorizer/gcc-loops -18.5309% +/- 6.8496% regressions: MultiSource/Benchmarks/ASCI_Purple/SMG2000/smg2000 18.351% +/- 12.156% SingleSource/Benchmarks/Shootout-C++/methcall 27.3086% +/- 14.4733% llvm-svn: 197099
2013-12-12 08:19:11 +08:00
int getOperandLatency(const InstrItineraryData *ItinData,
const MachineInstr *DefMI, unsigned DefIdx,
const MachineInstr *UseMI,
unsigned UseIdx) const override;
Improve instruction scheduling for the PPC POWER7 Aside from a few minor latency corrections, the major change here is a new hazard recognizer which focuses on better dispatch-group formation on the POWER7. As with the PPC970's hazard recognizer, the most important thing it does is avoid load-after-store hazards within the same dispatch group. It uses the POWER7's special dispatch-group-terminating nop instruction (instead of inserting multiple regular nop instructions). This new hazard recognizer makes use of the scheduling dependency graph itself, built using AA information, to robustly detect the possibility of load-after-store hazards. significant test-suite performance changes (the error bars are 99.5% confidence intervals based on 5 test-suite runs both with and without the change -- speedups are negative): speedups: MultiSource/Benchmarks/FreeBench/pcompress2/pcompress2 -0.55171% +/- 0.333168% MultiSource/Benchmarks/TSVC/CrossingThresholds-dbl/CrossingThresholds-dbl -17.5576% +/- 14.598% MultiSource/Benchmarks/TSVC/Reductions-dbl/Reductions-dbl -29.5708% +/- 7.09058% MultiSource/Benchmarks/TSVC/Reductions-flt/Reductions-flt -34.9471% +/- 11.4391% SingleSource/Benchmarks/BenchmarkGame/puzzle -25.1347% +/- 11.0104% SingleSource/Benchmarks/Misc/flops-8 -17.7297% +/- 9.79061% SingleSource/Benchmarks/Shootout-C++/ary3 -35.5018% +/- 23.9458% SingleSource/Regression/C/uint64_to_float -56.3165% +/- 25.4234% SingleSource/UnitTests/Vectorizer/gcc-loops -18.5309% +/- 6.8496% regressions: MultiSource/Benchmarks/ASCI_Purple/SMG2000/smg2000 18.351% +/- 12.156% SingleSource/Benchmarks/Shootout-C++/methcall 27.3086% +/- 14.4733% llvm-svn: 197099
2013-12-12 08:19:11 +08:00
int getOperandLatency(const InstrItineraryData *ItinData,
SDNode *DefNode, unsigned DefIdx,
SDNode *UseNode, unsigned UseIdx) const override {
Improve instruction scheduling for the PPC POWER7 Aside from a few minor latency corrections, the major change here is a new hazard recognizer which focuses on better dispatch-group formation on the POWER7. As with the PPC970's hazard recognizer, the most important thing it does is avoid load-after-store hazards within the same dispatch group. It uses the POWER7's special dispatch-group-terminating nop instruction (instead of inserting multiple regular nop instructions). This new hazard recognizer makes use of the scheduling dependency graph itself, built using AA information, to robustly detect the possibility of load-after-store hazards. significant test-suite performance changes (the error bars are 99.5% confidence intervals based on 5 test-suite runs both with and without the change -- speedups are negative): speedups: MultiSource/Benchmarks/FreeBench/pcompress2/pcompress2 -0.55171% +/- 0.333168% MultiSource/Benchmarks/TSVC/CrossingThresholds-dbl/CrossingThresholds-dbl -17.5576% +/- 14.598% MultiSource/Benchmarks/TSVC/Reductions-dbl/Reductions-dbl -29.5708% +/- 7.09058% MultiSource/Benchmarks/TSVC/Reductions-flt/Reductions-flt -34.9471% +/- 11.4391% SingleSource/Benchmarks/BenchmarkGame/puzzle -25.1347% +/- 11.0104% SingleSource/Benchmarks/Misc/flops-8 -17.7297% +/- 9.79061% SingleSource/Benchmarks/Shootout-C++/ary3 -35.5018% +/- 23.9458% SingleSource/Regression/C/uint64_to_float -56.3165% +/- 25.4234% SingleSource/UnitTests/Vectorizer/gcc-loops -18.5309% +/- 6.8496% regressions: MultiSource/Benchmarks/ASCI_Purple/SMG2000/smg2000 18.351% +/- 12.156% SingleSource/Benchmarks/Shootout-C++/methcall 27.3086% +/- 14.4733% llvm-svn: 197099
2013-12-12 08:19:11 +08:00
return PPCGenInstrInfo::getOperandLatency(ItinData, DefNode, DefIdx,
UseNode, UseIdx);
}
bool hasLowDefLatency(const TargetSchedModel &SchedModel,
const MachineInstr *DefMI,
unsigned DefIdx) const override {
// Machine LICM should hoist all instructions in low-register-pressure
// situations; none are sufficiently free to justify leaving in a loop
// body.
return false;
}
bool useMachineCombiner() const override {
return true;
}
/// Return true when there is potentially a faster code sequence
/// for an instruction chain ending in <Root>. All potential patterns are
/// output in the <Pattern> array.
bool getMachineCombinerPatterns(
MachineInstr &Root,
SmallVectorImpl<MachineCombinerPattern> &P) const override;
bool isAssociativeAndCommutative(const MachineInstr &Inst) const override;
bool isCoalescableExtInstr(const MachineInstr &MI,
unsigned &SrcReg, unsigned &DstReg,
unsigned &SubIdx) const override;
unsigned isLoadFromStackSlot(const MachineInstr *MI,
int &FrameIndex) const override;
unsigned isStoreToStackSlot(const MachineInstr *MI,
int &FrameIndex) const override;
bool findCommutedOpIndices(MachineInstr *MI, unsigned &SrcOpIdx1,
unsigned &SrcOpIdx2) const override;
void insertNoop(MachineBasicBlock &MBB,
MachineBasicBlock::iterator MI) const override;
// Branch analysis.
bool AnalyzeBranch(MachineBasicBlock &MBB, MachineBasicBlock *&TBB,
MachineBasicBlock *&FBB,
SmallVectorImpl<MachineOperand> &Cond,
bool AllowModify) const override;
unsigned RemoveBranch(MachineBasicBlock &MBB) const override;
unsigned InsertBranch(MachineBasicBlock &MBB, MachineBasicBlock *TBB,
MachineBasicBlock *FBB, ArrayRef<MachineOperand> Cond,
DebugLoc DL) const override;
// Select analysis.
bool canInsertSelect(const MachineBasicBlock &, ArrayRef<MachineOperand> Cond,
unsigned, unsigned, int &, int &, int &) const override;
void insertSelect(MachineBasicBlock &MBB, MachineBasicBlock::iterator MI,
DebugLoc DL, unsigned DstReg, ArrayRef<MachineOperand> Cond,
unsigned TrueReg, unsigned FalseReg) const override;
void copyPhysReg(MachineBasicBlock &MBB,
MachineBasicBlock::iterator I, DebugLoc DL,
unsigned DestReg, unsigned SrcReg,
bool KillSrc) const override;
void storeRegToStackSlot(MachineBasicBlock &MBB,
MachineBasicBlock::iterator MBBI,
unsigned SrcReg, bool isKill, int FrameIndex,
const TargetRegisterClass *RC,
const TargetRegisterInfo *TRI) const override;
void loadRegFromStackSlot(MachineBasicBlock &MBB,
MachineBasicBlock::iterator MBBI,
unsigned DestReg, int FrameIndex,
const TargetRegisterClass *RC,
const TargetRegisterInfo *TRI) const override;
bool
ReverseBranchCondition(SmallVectorImpl<MachineOperand> &Cond) const override;
bool FoldImmediate(MachineInstr *UseMI, MachineInstr *DefMI,
unsigned Reg, MachineRegisterInfo *MRI) const override;
// If conversion by predication (only supported by some branch instructions).
// All of the profitability checks always return true; it is always
// profitable to use the predicated branches.
bool isProfitableToIfCvt(MachineBasicBlock &MBB,
unsigned NumCycles, unsigned ExtraPredCycles,
BranchProbability Probability) const override {
return true;
}
bool isProfitableToIfCvt(MachineBasicBlock &TMBB,
unsigned NumT, unsigned ExtraT,
MachineBasicBlock &FMBB,
unsigned NumF, unsigned ExtraF,
BranchProbability Probability) const override;
bool isProfitableToDupForIfCvt(MachineBasicBlock &MBB, unsigned NumCycles,
BranchProbability Probability) const override {
return true;
}
bool isProfitableToUnpredicate(MachineBasicBlock &TMBB,
MachineBasicBlock &FMBB) const override {
return false;
}
// Predication support.
bool isPredicated(const MachineInstr &MI) const override;
bool isUnpredicatedTerminator(const MachineInstr &MI) const override;
bool PredicateInstruction(MachineInstr &MI,
ArrayRef<MachineOperand> Pred) const override;
bool SubsumesPredicate(ArrayRef<MachineOperand> Pred1,
ArrayRef<MachineOperand> Pred2) const override;
bool DefinesPredicate(MachineInstr &MI,
std::vector<MachineOperand> &Pred) const override;
bool isPredicable(MachineInstr &MI) const override;
// Comparison optimization.
bool analyzeCompare(const MachineInstr *MI,
unsigned &SrcReg, unsigned &SrcReg2,
int &Mask, int &Value) const override;
bool optimizeCompareInstr(MachineInstr *CmpInstr,
unsigned SrcReg, unsigned SrcReg2,
int Mask, int Value,
const MachineRegisterInfo *MRI) const override;
/// GetInstSize - Return the number of bytes of code the specified
/// instruction may be. This returns the maximum number of bytes.
///
unsigned GetInstSizeInBytes(const MachineInstr *MI) const;
void getNoopForMachoTarget(MCInst &NopInst) const override;
std::pair<unsigned, unsigned>
decomposeMachineOperandsTargetFlags(unsigned TF) const override;
ArrayRef<std::pair<unsigned, const char *>>
getSerializableDirectMachineOperandTargetFlags() const override;
ArrayRef<std::pair<unsigned, const char *>>
getSerializableBitmaskMachineOperandTargetFlags() const override;
// Lower pseudo instructions after register allocation.
bool expandPostRAPseudo(MachineBasicBlock::iterator MI) const override;
};
}
#endif